EEMD and combined kernel RVM-based photovoltaic power short-term prediction method

Abstract

The invention discloses an EEMD (Ensemble Empirical Mode Decomposition) and combined kernel RVM (Relevance Vector Machines)-based photovoltaic power short-term prediction method. Through the adopted EEMD, a problem that the possibility of mode mixing during empirical mode decomposition is high is avoided, so that higher resolution and very strong non-linear processing capacity are provided, the data complexity is reduced, the effect is excellent, the result is accurate, the prediction accuracy is effectively improved, and the method can be applied to the data preprocessing of the photovoltaic output power well; through the adoption of an RVM method for short-term photovoltaic power prediction, the method has the advantages of being sparse in model height, less in kernel parameters to be optimized, flexible in kernel function selection, and high in model generalization ability; and by use of a combined kernel function, the prediction accuracy of photovoltaic power through a model is further improved when the weather changes suddenly, so that the universal adaptability and the generalization performance of the model are improved.

Description

technical field [0001] The invention belongs to the technical field of new energy power generation and smart grid, and specifically relates to a short-term prediction method of photovoltaic power based on EEMD and combined core RVM. Background technique [0002] After the 1970s, with the development of industrialization, fossil fuels are facing depletion, and environmental problems have become increasingly prominent. In order to solve this problem, human beings have begun to pay attention to renewable energy, among which solar energy has become the focus of everyone's attention. By the end of 2014, the installed capacity of photovoltaic power plants in China will reach 14GW, and it is estimated that the installed capacity of photovoltaic power plants will reach 100-200GW in 2030. However, photovoltaic power generation is easily disturbed by many climatic factors. When the power disturbance after grid connection is serious, it may affect the safe and stable operation of the ...

AI Technical Summary

Problems solved by technology

[0005] However, the problems of information processing methods such as wavelet analysis, EMD and LMD are: artificial setting is required, subjectivity is strong, and empirical mode decomposition is prone to mode aliasing

However, in the actual application process, because the time series does not consider the influence of external factors for the time being, when the external environment changes greatly, the prediction error is often large; the ANN method is easy to lead to the problem of insufficient learning or over-fitting during training;

[0007] Although machine learning algorithms such as SVM can effectively avoid the risk of falling into a local minimum and achieve more accurate predictions, they still have the following shortcomings: ①The kernel function must meet the Mercer condition, and there are few optional kernel functions; ②There are many parameters and support The vector increases linearly with the increase of training samples, and the amount of calculation is large; ③When there are many input influencing factors, the structure of the prediction model will be too complex and the training efficiency will be low

Images